Project Summary/Abstract The proteome is a quantitative output of a genome and the ultimate effector of cellular functions. Yet remarkably little is known about the logic behind proteome construction. The goal of my research program is to understand the evolutionary driving forces that shape protein levels, as well as the precise regulation that is employed to arrive at the exact set point. Our entry point is a powerful quantitative proteomics method based on ribosome profiling. Using bacterial model systems, we have shown that many homologous proteins have quantitatively conserved expression levels across divergent species, suggesting strong constraints on protein abundance that we do not currently understand. To elucidate the mechanistic basis for the preferred protein levels in relation to cell fitness, we are using theory-guided experimental design to investigate the consequence of protein misregulation on global physiology in E. coli and B. subtilis. Furthermore, we are establishing a comprehensive comparative proteomics approach to broadly identify key proteins whose levels are invariant despite regulatory changes throughout evolution. This approach will open up new avenues for uncovering the control points of biochemical systems, and provide a new way of thinking about proteome imbalance in diseases. The second arm of my research program is to understand how cells produce their proteome with quantitative precision. As indicated by our preliminary results, such precise control is clearly important to many proteins, but we have only a rudimentary understanding of how these rates are finely tuned. We are now developing genome-wide techniques to interrogate the prevalence of feedback regulation in maintaining proteome homeostasis in bacteria. We are also investigating how co-regulated genes in operons are differentially expressed to achieve exact protein levels. Our approach in this area focuses on the development of quantitative assays that allow us to accurately characterize the molecular processes perfected by natural selection. We anticipate that our mechanistic dissection, coupled with systems-level inquiry into proteome composition, will make bacterial model organisms the first system for which we have a quantitative understanding from genome to proteome to fitness and back.